niamh dunne prepared for eusa conference, 3-5 march 2011
TRANSCRIPT
1
Margin Squeeze: Theory, Practice, Policy
Niamh Dunne1
Prepared for EUSA Conference, 3-5 March 20112
ABSTRACT: Margin squeeze occurs where the margin between the price charged by a vertically integrated firm for a wholesale input, and its own retail price for the end product incorporating the input, is so low as to foreclose one or more affected markets. The extent to which margin squeeze should constitute a discrete competition law offence, distinct from predation or refusal to deal, is a disputed question. A jurisprudential chasm between the approaches to margin squeeze under European Union competition law and United States antitrust has emerged, following the Court of Justice of the European Union’s judgments in Deutsche Telekom and TeliaSonera and the US Supreme Court’s decision in LinkLine. The EU recognises a broad concept of margin squeeze, applicable in any sector; the US does not recognise margin squeeze as a standalone abuse, and moreover, the presence of sector-specific regulation excludes the application of antitrust to the price levels that comprise the squeeze. This paper explores the margin squeeze concept, with particular attention to both areas of contention.
Protection of competition not competitors is the accepted wisdom behind the competition rules. To
what extent, however, may a vertically integrated dominant firm be liable for failure to protect its
competitors’ profit margins, where this in turn harms competition? That is, in effect, the problem
posed by margin squeeze. This paper examines the margin squeeze concept and its relationship to
sector-specific economic regulation in three dimensions: theory, practice and policy implications.
The paper is structured as follows. Part II sets out the basic theory of margin squeeze; explains how
such a practice may raise problems from a competition perspective; and describes how the abuse has
been accommodated under the competition rules. Part III examines the case law on margin squeeze
within the European Union (EU), from its first mention in the Napier Brown Commission decision, to
the recent Court of Justice judgments in the Deutsche Telekom appeal and the TeliaSonera reference
case. Given the apparently opposite answers of United States courts when faced with broadly similar
questions, the US jurisprudence on the issue of margin squeeze is also considered. Part IV of the
paper considers the impact of sector-specific regulation on margin squeeze as a principle of
competition law, considering in particular the policy issues that arise from the concurrency approach
adopted under current EU law. Part V of the paper concludes. 1 PhD candidate in law, University of Cambridge, [email protected]. 2 This is a WORK IN PROGRESS and any comments would be most gratefully received. Please do not cite or circulate without permission.
2
II: The Theory of Margin Squeeze
Margin squeeze is a problem that can arise where a firm is organised with a degree of vertical
integration, meaning that it operates at two or more levels of the supply chain. Microeconomic theory
posits that economic operators vertically integrate in circumstances where this is the most efficient
manner in which to organise their businesses, with the result that consumer welfare as a whole is
maximised. Thus, vertical integration is not an a priori problem, and indeed generally brings
beneficial efficiencies of scale and scope to the firm’s operations.3
Where, however, the vertically integrated firm sells part of its wholesale level production to a non-
integrated firm that competes with it at the downstream level, the integrated firm gains a measure of
control over its competitor’ costs—which can be significant, particularly where there is no alternative
source of supply for the input available to the downstream competitor, for example where there is a
“bottleneck monopoly” at the wholesale level. Moreover, where the integrated firm also holds a
dominant market position downstream, its ability to influence prices in that market means that it can
also exert control over its competitor’s revenues. Thus, the integrated firm may have the ability to
manipulate, or “squeeze”, the relationship between the competitor’s costs and revenues in order to
reduce its profit margin to such a low (or negative) level as to force the competitor from the market.4
Figure 1 below illustrates this vertical price relationship in graphical form.
3 The classic exposition of this theory is found in R.H. Coase, “The Nature of the Firm” (1937) 4(16) Economica 386-405. From a competition law perspective, a key benefit that is claimed about vertical integration is that it prevents double marginalisation, that is, the integrated firm can earn monopoly profit at only one level of the supply chain. By contrast, where there is a vertically separated market structure, monopoly profits can be made at each level of the supply chain, and consumer welfare as a whole is diminished. For a relatively neutral discussion of the double marginalisation issue, see W.K. Viscusi, J.E. Harrington & J.M. Vernon (2005), Economics of Regulation and Antitrust 4th ed., MIT Press, Cambridge, Massachusetts, USA, in particular Chapter 8: “Vertical Mergers and Vertical Restraints”. While the “one monopoly profit” theory—a lynchpin of the Chicago School approach to competition law—has come in for criticism in more recent years, nonetheless there is considerable evidence that vertical integration can result in significant welfare benefits: see F. Lafontaine & M Slade, “Vertical Integration and Firm Boundaries: The Evidence” (2007) 45(3) Journal of
Economic Literature 629-685 for a wide-ranging review of studies on the impact of vertical integration, concluding that under most circumstances, profit-maximising vertical integration decisions are efficient, from the perspective of both firms and consumers. 4 This explanation, based on the textbook example of margin squeeze, is presented in a simplified form and is intended to illustrate the legal problem under consideration rather than providing an authoritative statement of the economic issues concerned. See OECD, Roundtable on Margin Squeeze DAF/COMP(2009)36, at 24, for a neat summary of the assumptions of the classic margin squeeze case, as well as a sample of differences that arise in real world scenarios. See also D. Geradin & R. O’Donoghue, “The Concurrent Application of Competition Law and Regulation: The Case of Margin Squeeze Abuses in the Telecommunications Sector” (2005) 1(2) Journal of Competition Law & Economics 355, at 358-360, for an outline of the basis conditions for margin squeeze.
3
Figure 1: Vertical Margin Squeeze
Does a margin squeeze harm competition in the market, or does it merely diminish the competing
firm’s profits? Critics of margin squeeze attack it as being “an antitrust rule that punishes a firm for
failing to ensure its competitors’ profitability.”5 They argue that prohibiting margin squeeze simply
leads to a wealth transfer from the vertically integrated firm to its downstream competitor, with no
increase in consumer welfare, and moreover, can have a negative impact on the investment and
competition incentives of the dominant firm. In addition, where the remedy for a squeeze is to
increase retail prices to consumers, as is in fact expressly permitted under EU law, consumers lose out
both directly as a result of higher prices and indirectly because of the negative impact on competition
in the marketplace.
Moreover, there are those who argue that a margin squeeze will never (or at least rarely) be a rational
strategy for an integrated firm: even though the firm would gain by acquiring more customers at the
retail level from excluding its rivals, it would still lose out because of reduced profits at the wholesale
5 J.G. Sidak, “Abolishing the Price Squeeze as a Theory of Antitrust Liability” (2008) 4(2) Journal of
Competition Law & Economics 279, at 294.
Vertically integrated (VI) firm selling
wholesale input
VI firm selling
retail product
Downstream (DS) competitor buying
wholesale input
DS competitor
selling retail
product
Potential for
margin
squeeze
between
wholesale and
retail price
levels
4
level, where it has lost the customers of its wholesale product. This argument is tied to the notion that
there is but “one monopoly profit” for the vertically integrated firm.6
Nonetheless, a plausible theory of competition harm (and corresponding benefit for the integrated
firm) resulting from margin squeeze can be constructed. Firstly, exclusion of the rival from the
downstream market reduces competition at that level, thus reducing the level of competitive pressure
on the integrated firm to maintain lower prices and/or higher product quality. For those who view the
maximisation of consumer welfare as the sole objective of competition law, a diminution in
competition may not be seen as a problem where it does not result in an overall diminution in
welfare.7 For those who see competition law as a tool by which to protect the competition process,
however, loss of competitors can greatly impair the proper functioning of the market mechanism.8
Secondly, even if the rival does not exit the downstream market, the squeeze may prevent it from
climbing the ladder of investment; that is, integrating upstream into the dominant firm’s primary
market.9 If the rival can develop its own upstream infrastructure, the bottleneck problem disappears,
negating the possibility that a margin squeeze can arise in future and probably the need for economic
regulation of the market. However, the existence of the margin squeeze prevents the rival from
having sufficient revenues to fund and/or sufficient downstream market share to justify such
investment.
Thirdly, the margin squeeze may also limit the rival’s incentive to innovate within the downstream
market, insofar as any cost saving the rival makes as a result of greater efficiency will be captured by
the dominant firm via the squeeze. This results in consumer harm because it prevents the emergence
of a more efficient downstream market.10
The Telefónica decision provides a good example of the related ways in which foreclosure via margin
squeeze can be a rational, profitable strategy for a dominant firm. In this case, the Commission found
that cross-subsidies from the high retail prices that Telefónica could maintain through foreclosure of
the downstream market surpassed by far the foregone profits at the retail level; Telefónica was able to
6 D.W. Carlton, “Should “Price Squeeze” be a Recognized Form of Anticompetitive Conduct?” (2008) 4(2) Journal of Competition Law & Economics 271, at 275. Geradin & O’Donoghue (2005), at 364, argue that while the reduced incentive for the firm to engage in margin squeeze does not render such behaviour irrational as a matter of course, it means that competition enforcers should consider carefully whether foreclosure is plausible in the circumstances. 7 OECD (2009) at 27. 8 See, for example, E.M. Fox, “The Modernization of Antitrust: A New Equilibrium” (1981) 66 Cornell Law
Review 1140 at 1191. 9 See E.N. Hovenkamp & H. Hovenkamp, “The Viability of Antitrust Price Squeeze Claims” (2009) 51 Arizona
Law Review 273, at 287. 10 Hovenkamp & Hovenkamp (2009), at 289.
5
protect its dominant position in adjacent retail markets; and it was perfectly position to also pre-empt
emerging retail markets in new technologies.11 At the same time, the lack of competition in the
downstream market meant that Spain had among the highest average prices for retail broadband in the
EU, and below-average broadband penetration rates. Thus, Telefónica maintained its dominance and
consumers suffered due to few users and higher prices.
Margin Squeeze and Competition Law
The principal focus of this paper is an analysis of the treatment of margin squeeze cases under current
competition jurisprudence, firstly in the EU, with a comparative assessment of the approach under US
antitrust law. Particular consideration is given to the question of the application of the margin
squeeze concept, as a principle of competition law, in sectors already subject to sector-specific
regulation. Before we consider the specifics of the approaches in the two jurisdictions under
examination, however, it is worthwhile giving some thought, more generally, to how price squeezing
behaviour fits within the established competition law framework.
Although a vertically integrated firm may, legally, comprise a number of separate companies, the
group of companies as a whole typically constitutes a single economic unit. Therefore, the
competition rules on unilateral conduct are the appropriate starting point for our assessment. In the
EU, Article 102 of the Treaty on the Functioning of the European Union (TFEU)12 prohibits the
“abuse by one or more undertakings of a dominant position within the internal market or in a
substantial part of it”, insofar as this affects trade between Member States. In the US, §2 of the
Sherman Act prohibits monopolisation, attempts to monopolise and conspiracies to monopolise trade.
While the market share threshold for the application of Article 102 TFEU is significantly lower than
that required for the application of §2 of the Sherman Act, the latter provision covers the
anticompetitive acquisition as well as maintenance of a monopoly, whereas at least according to the
letter of the law Article 102 TFEU covers only the abuse of an existing position of dominance. For
the purposes of this paper, however, we shall assume that the core of each provision is broadly
equivalent: prohibiting anticompetitive behaviour by firms holding substantial market power.
How might a margin squeeze be construed as a unilateral conduct abuse? Firstly, there are those that
would exclude circumstances leading to a margin squeeze from the purview of competition law
11 Commission Decision of 4 July 2007 relating to a proceeding under Article 83 of the EC Treaty (Case COMP/38.784 – Wanadoo España v Telefónica), at paragraphs 611-613. 12 Prior to the entry into force of the Lisbon Treaty on 1 December 2009, Article 102 TFEU was known as Article 82 EC, and prior to that Article 86 of the EC Treaty. For the sake of simplicity, the term “Article 102 TFEU” is used throughout this paper to refer to the provision, which remains completely unchanged in substance.
6
entirely. Rather, the argument is that such behaviour is more appropriately dealt with under the
economic principles on access pricing. Thus, the wholesale level price (and possibility also the
downstream price) should be set by a regulator at a level that maximises consumer welfare, the
assumption being that in the market concerned the operation of the market mechanism will not result
in a socially optimal pricing scheme.13 A key consideration for those favouring the regulatory
approach is the presumed superiority of regulators in addressing competition issues within their area
of special expertise, when compared to generalist competition law enforcers.14 On the other hand,
sector-specific regulation and competition law seek to achieve different objectives with respect to
competition—broadly, promoting versus policing competition, respectively—a distinction that may
justify a different economic treatment of margin squeeze under each approach,15 and which therefore
suggests that subsuming antitrust considerations within access pricing is not the optimum solution.
Secondly, even where it is admitted that margin squeeze behaviour may be examined under the
competition rules, opinion is divided as to whether a margin squeeze should constitute a standalone
cause of action, or whether it should instead be fitted into existing competition law standards. The
latter position is supported by the argument that competition law should treat economically-equivalent
actions in an identical manner, and that many if not all forms of conduct by which a margin squeeze
can arise are already captured by existing competition law prohibitions.16
Thus, some commentators treat margin squeeze as a form of refusal to deal,17 an approach that focuses
principally on the wholesale prices charged by the vertically integrated firm. Under this view, the
dominant firm’s practice of charging a wholesale prices that does not allow the downstream
competing firms to make sufficient profit (however one defines sufficiency) in view of the existing
retail price amounts to a constructive refusal to deal, or a refusal to deal on commercially acceptable
terms. As an competition law abuse, however, refusal to deal is construed very narrowly—the default
assumption being that even dominant firms are free to decide who they want to transact with and on
what terms—so that an approach to margin squeeze that examines the behaviour solely through the
lens of the refusal to deal principle is in practice a very restricted one.
13 For example, see Sidak (2008); also OECD (2009), cited fn. 4 above, at 27-29, for a summary of access pricing principles. 14 See Sidak (2008), at 287 & 294 and Hovenkamp & Hovenkamp (2009) at 297-298. 15 See Oxera, “No Margin for Error: The Challenges of Assessing Margin Squeeze in Practice” Oxera Agenda,
November 2009. 16 OECD (2009) at 21. 17 For example, S.C. Salop, “Refusals to Deal and Price Squeezes by an Unregulated, Vertically Integrated Monopolist” (2010) 76(3) Antitrust Law Journal 709 and E. Meriwether, “Putting the “Squeeze” on Refusal to Deal Cases: Lessons from Trinko and LinkLine” (2010) 24 Antitrust 65.
7
Alternatively, margin squeeze may be characterised as a form of predation, that is, below cost pricing,
sustained by the dominant firm for long enough to drive the competitor from the market.18 Colley &
Burnside favour this approach on the basis that, under both predation and margin squeeze, the
dominant firm incurs incremental losses (in the latter case, at least in the form of profits forgone).19
As the Telefónica case demonstrates, however, margin squeeze cases can just as easily involve high
wholesale and retail prices as low wholesale and retail levels. Moreover, the test for predatory pricing
test, which assumes that predation is an inherently unsustainable policy insofar as no firm can lose
money forever, may not capture fully the vertical element of margin squeeze. Vertical integration and
cross-subsidisation means that margin squeeze contains a built-in mechanism by which the integrated
firm can sustain the otherwise unsustainable.20
It is this vertical aspect of margin squeeze that prompts other commentators to argue that it should
constitute a standalone abuse, on the basis that examination of each price level separately for evidence
of a discrete abuse fails to take account of the dominant firm’s ability to engage in vertical
leveraging.21 The issue then becomes one of determining what level of a squeeze between wholesale
and retail prices should be prohibited—how “unfair”, “inadequate”, “inappropriate” or
“disproportionate” the relationship between these prices must be,22 and more importantly, how this is
to be measured, objectively and efficiently.23
Some would go beyond even an economics-based standalone test for margin squeeze. Lopatka, who
was in fact sceptical of the logic of a quantitative margin squeeze test in the presence of price
regulation, argued that dominant firms should instead be liable for deliberate abusive conduct in the
regulatory price-setting process. Conversely, a margin squeeze that arises as a result of regulatory
inconsistency would not be actionable.24 On the other hand, Dogan & Lemley, who have argued that
18 For example, Carlton (2008); S.L. Dogan & M.A. Lemley, “Antitrust Law and Regulatory Gaming” (2009) 87 Texas Law Review 685, at 725; and L Colley & S. Burnside, “Margin Squeeze Abuse” (2006) 2 European
Competition Journal 185, at 186-188. Sidak (2008), although favouring a access pricing approach, argues that where the retail price is unregulated its legality should be considered under the predatory pricing rules. 19 Colley & Burnside (2006) at 186. 20 Hovenkamp & Hovenkamp (2009), at 291. 21 N. Economides, “Vertical Leverage and the Sacrifice Principle: Why the Supreme Court got Trinko Wrong” (2005) 61 NYU Annual Survey of American Law 379. 22 To use the language of the TeliaSonera, LinkLine, Deutsche Telekom (CJEU) judgments and the Telefónica decision, all discussed in Part III below, respectively. 23 Two options are the as-efficient competitor test, which uses the costs of the dominant firm, and the reasonably efficient competitor test, which uses the costs of a hypothetical reasonably efficient competitor of the dominant firm. 24 J.E. Lopatka, “The Electric Utility Price Squeeze as an Antitrust Cause of Action” (1984) 31 UCLA Law
Review 563. Note, however, that Lopatka was considering circumstances where the wholesale and retail prices were set by separate regulatory agencies. Where both price-setting functions reside in a single agency, he suggested that participation in the regulatory process would probably be immunised by the Noerr-Pennington doctrine of US antitrust law, which excludes lobbying and other attempts to influence the legislative function from antitrust enforcement (at 634). The case theory behind the prosecution of Rambus by both the European
8
competition law should be used to address other forms of regulatory gaming such as patent hopping,
see margin squeeze as the limit to this approach: where there is detailed regulation already in place,
the costs of antitrust intervention begin to outweigh its benefits.25
This latter point leads to an additional question: even where we recognise margin squeeze as a
competition law abuse (either as a standalone offence or where the component prices constitute
discrete examples of other established offences such as predation or refusal to deal), to what extent
does the existence of ex ante sector-specific economic regulation in the market impact upon, or even
prevent, the ex post application of competition law to the price squeeze behaviour? This question is
linked to broader policy considerations regarding the intersection between competition law and
economic regulation more generally, but it has a particular significance in the margin squeeze context.
As noted, many margin squeeze cases arise in sectors where there is a bottleneck monopoly at the
wholesale level—and it is the presence of such a monopoly that is, frequently, the impetus behind the
implementation of economic regulation in the first place. Thus, arguably, margin squeeze cases are in
practice most likely to arise in sectors where there is existing regulation, meaning that the relationship
between margin squeeze as a competition law and regulatory abuse merits special consideration.
Bouckaert & Verboven, for example, argue that the presence, and degree, of economic regulation
should impact on the treatment of margin squeeze on the basis that the regulatory environment
determines the nature of the offence.26 Where there is full regulation at both the wholesale and retail
levels, they argue that a squeeze is simply “an artefact of the regulatory system”,27 and therefore can
only be addressed through the regulation in place. Where there is price regulation at the wholesale
but not the retail level, competition law can be applied to the unregulated retail price in the form of
the test for predatory pricing, but not to the margin as such. It is only where there is no regulation at
either level that foreclosure via the margin itself becomes a problem—but even in such circumstances,
competition law enforcement may not be the optimal tool to resolve the problem. Rather, it may be
better addressed through alternative solutions such as (re)regulation, facilities-based competition
(although this is not always economically feasible) or structural separation of the vertically integrated
firm.28 While this approach is merely one among many to the regulation/competition law interface in
Commission and the US Federal Trade Commission, concerning anticompetitive participation in a standard-setting process, might provide some support for this approach, but unfortunately neither proceeding has resulted in a formal finding of breach: the EU proceedings were closed without a finding of abuse on receipt of behavioural commitments from the firm; the initial finding of breach by the FTC was annulled on appeal. 25 Dogan & Lemley (2009) at 723. 26 J. Bouckaert & F. Verboven, “Price Squeezes in a Regulatory Environment” (2004) 26(3) Journal of
Regulatory Economics 321. 27 Bouckaert & Verboven (2004), at 334. 28 Structural separation of vertically integrated firms can provide a more enduring remedy to margin squeeze problems by separating the upstream and downstream activities of the firm, effectively disintegrating the integrated firm. It is an interesting (and controversial) topic beyond the remit of this paper. Detailed discussion
9
the margin squeeze context, it does illustrate the significant impact that regulation may have on both
the policy and economics-focused issues. This aspect will be discussed in detail in Part IV of the
paper.
III. Margin Squeeze: The Case Law
Having considered the theory of margin squeeze, we shall now examine the development of the EU
competition law jurisprudence on the issue, as well as outlining the current position. The EU
approach is compared to and contrasted with the far more restrictive US approach, a divergence that is
particularly interesting in view of the close factual similarities between the cases that have arisen in
both jurisdictions. In particular, we shall examine two issues: (i) the extent to which margin squeeze
constitutes a standalone competition law offence; and (ii) the impact of sector-specific regulation on
margin squeeze as a competition law concept. In both jurisdictions, as noted above, we are
considering the position of the vertically integrated form under the rules relating to unilateral conduct,
that is Article 102 TFEU and §2 of the Sherman Act.
Margin Squeeze in the European Union
The concept of margin squeeze as an abuse of dominance was first applied by the Commission in
1988 in its Napier Brown – British Sugar decision.29 The case concerned a series of abusive practices
carried out by the sole sugar beet producer in the UK, British Sugar. This firm was active, inter alia,
on the markets for the sale of granulated sugar in bulk quantities and individually packaged for retail
sale in the UK, and was found by the Commission to hold a dominant position on both markets.
When the UK’s largest sugar merchant, Napier Brown, attempted to enter the retail packet sugar
market, by buying sugar in bulk from British Sugar and packaging itself, British Sugar retaliated with
a series of sales practices designed to prevent Napier Brown from gaining a foothold in the
downstream market, including refusals to supply, tying and fidelity rebates. Among the alleged
anticompetitive practices was the fact that British Sugar had cut its prices for retail packet sugar,
thereby squeezing the margin between the wholesale cost of sugar in bulk and retail packet sugar
prices. The Commission took the view that a margin squeeze of this nature could amount to an abuse
of dominance, in the circumstances:
of the policy considerations involved can be found in OECD, Restructuring Public Utilities for Competition (2001). A useful description of the degrees of separation that may be imposed is available in M. Cave, “Six Degrees of Separation: Operation Separation as a Remedy in European Telecommunications Regulation” 64 Communications and Strategies (4th quarter 2006), p.1-15. For an example of recent economic thinking on structural separation, see B. Moselle & D. Black, “Vertical Separation as an Appropriate Remedy” (2011) 2(1) Journal of Competition Law & Practice 84. 29 Commission Decision 88/518/EEC of 18 July 1988 relating to a proceeding under Article 86 of the EEC Treaty (Case No IV/30.178 Napier Brown – British Sugar) (OJ L 284/41, 19.10.1988).
10
The maintaining, by a dominant company, which is dominant in the markets for both a raw material and a corresponding derived product, of a margin between the price which it charges for a raw material to the companies which compete with the dominant company in the production of the derived product and the price which it charges for the derived product, which is insufficient to reflect that dominant company’s own costs of transformation…[]…with the result that competition in the derived product is restricted, is an abuse of dominant position.
30
Thus, the Commission established that the relevant measure of costs is an objective one: the as-
efficient competitor test, that is, British Sugar’s own costs, rather than those of its downstream
competitors. The standard by which those costs were to be assessed, however, was a rather nebulous
one: the sufficiency of the margin in reflecting the cost of transformation between the wholesale and
retail products. On the facts, taking into account the evidence of intention to foreclose the retail sugar
market, and coupled with the other abuses that British Sugar has committed, the Commission held that
the margin squeeze constituted an abuse of dominance.31
The possibility that a margin squeeze can amount to an abuse of dominance was later recognised by
the then Court of First Instance (now General Court) in the Industries des Poudres Sphériques
judgment,32 albeit no abuse was found on the circumstances. Industries des Poudres Sphériques (IPS)
was a French firm that manufactured that manufactured broken calcium metal from primary calcium
metal. It purchased the latter raw material from another French firm, Péchiney Électrométallurgie
(PEM), which was also active in the broken calcium metal market. IPS complained to the
Commission, inter alia, that PEM’s pricing practices with respect to its primary calcium metal and
broken calcium metal products, respectively, constituted an abuse of PEM’s alleged dominant
position on the upstream market. The Commission rejected the complaint against PEM in its entirety,
and IPS appealed the rejection decision to the General Court. In assessing the initial complaint
submitted by IPS, the General Court described the potential abuse of margin squeeze in the following
terms:
Price squeezing may be said to take place when an undertaking which is in a dominant position on the market for an unprocessed product and itself uses part of its production for the manufacture of a more processed product, while at the same time selling off surplus unprocessed product on the market, sets the price at which it sells the unprocessed product at such a level that those who purchase it do not have a sufficient profit margin on the processing to remain competitive on the market for the processed product.33
30 Napier Brown at paragraph 66, relying on the earlier Court of Justice decision in Case 6/72 Continental Can v
Commission [1973] ECR 215. 31 The other abuses challenged in the decision contributed to the margin squeeze found on the facts. British Sugar refused to supply bulk sugar apart from on an ex factory pricing basis, which included a charge for the delivery of the sugar from British Sugar’s factory to the customer’s premises. Not only did this practice have the effect of reserving to British Sugar the separate but ancillary activity of delivery of the sugar, it also raised the wholesale cost of the sugar for any firm intending to transform the bulk sugar into packet sugar. 32 Case T-5/97Industries des Poudres Sphériques v Commission [2000[ ECR II-3755. 33 Industries des Poudres Sphériques at paragraph 178.
11
Thus, the Genera; Court followed the Commission’s approach in Napier Brown in adopting the
sufficiency of the margin as the central concern. Moreover, by focusing on the costs and revenues of
the downstream competitor, rather than the dominant firm, the formulation in Industries des Poudres
Sphériques appears, at first glance, to be more favourable to downstream competitors than the as-
efficient competitor test, insofar as it protects inefficient competitors in addition to as-efficient ones.
Nonetheless, when it came to applying this formulation to the facts at hand, the CFI adopted a far
more restrictive approach. It held that, in the absence of abusive prices at the wholesale level (which
would constitute a constructive refusal to deal) or predatory prices at the retail level, IPS had failed to
raise facts that could constitute a breach of the competition rules. In particular, the fact that IPS
incurred higher processing costs did not justify characterising PEM’s pricing policy as abusive—
because a producer, even in a dominant position, is not obliged to sell its product below its
manufacturing costs.34
Another case of relevance in this context, albeit one that does not consider the issue of vertical margin
squeeze directly, is the Deutsche Post decision of 2001.35 In this case, the Commission held that
Deutsche Post, the incumbent postal operator in Germany, had abused its dominant position in the
liberalised market for mail order parcel services in Germany, by setting predatory prices for those
services that did not cover the incremental cost of their provision. While the Commission applied the
orthodox predatory pricing standards established in AKZO,36 the predation strategy was financed via
cross-subsidies from the profitable letter-post market, where Deutsche Post held a legal monopoly of
service. While the case thus concerned horizontal rather than vertical related markets, it provides
support for the proposition that a firm’s ability to cross-subsidise between profitable and losing-
making businesses can make a foreclosure strategy rational in the longer term.37
Nonetheless, until recently the parameters of the margin squeeze concept as a unilateral conduct
offence were somewhat unsettled within EU competition law: firstly, as to whether it can constitute a
standalone competition abuse, in the absence of predation or refusal to deal; and secondly, as to the
calculation of whether a margin is abusive on the facts. These questions have been substantially
clarified in two recent judgments of the Court of Justice of the European Union dealing expressly and
34 Industries des Poudres Sphériques at paragraph 179. Note that, while IPS had higher processing costs than PEM, it ended up with a higher value final product, insofar as the broken calcium metal sold by IPS commanded a 25% premium on the prices charged for competing products; see Industries des Poudres Sphériques at paragraph 185. 35 Commission Decision 2001/354/EC of 20 March 2001 relating to a proceeding under Article 82 of the EC Treaty (Case Comp/35.141 – Deutsche Post AG) (OJ L 125/27, 5.5.2001). 36 Case C-62/86 AKZO v Commission [1991] ECR I-3359. 37 See G.R. Faulhaber, “Cross-Subsidization: Pricing in Public Enterprises” (1975) 65(5) American Economic
Review 966-977 for a discussion on cross-subsidisation by integrated firms, in particular the impact on consumer welfare and the possibility of competitive entry.
12
at length with the issue of margin squeeze, namely Deutsche Telekom and TeliaSonera. In these
cases, the Court has recognised a very broad concept of margin squeeze as a standalone abuse, and
moreover on that can arise in any circumstances, whether in a regulated sector, in the presence of an
essential facility or in an otherwise competitive market.
The Deutsche Telekom Judgment
The Deutsche Telekom case emerged from the EU’s efforts to introduce competition at the retail level
for telecommunications services in the EU, via mandatory unbundling requirements imposed on
incumbent telecommunications operators in the Member States. Deutsche Telekom (DT) is the
formerly State-owned incumbent operator in Germany, which retains ownership of the fixed
telephone network. The German markets in the provision of infrastructure and telephone services
have been liberalised since 1996, in line with EU law requirements, and DT has been subject to
mandatory access requirements under national law since that date. DT now provides, inter alia, retail
telecommunications services to end users, as well as wholesale access to infrastructure services to
non-integrated firms that compete with it in the downstream markets. During the time period in
question, in addition to the mandatory access requirements, DT was subject to price regulation at both
the wholesale and retail levels. The case against DT proceeded on the basis that, at the wholesale
level, access charges were set by the national telecommunications regulator, while at the retail level,
DT was subject to a price cap for “baskets” of services.38 The latter regulatory structure required that
the aggregate price for each basket was not permitted to exceed a certain level, but gave DT discretion
as to the pricing of individual component services within the basket, although any adjustment of these
charges required authorisation from the regulator.
By a decision of 2003, the Commission held that the spread between the wholesale and retail prices
charged by DT for fully unbundled access to its local loops amounted to an abusive margin squeeze,
contrary to what is now Article 102 TFEU.39 DT was considered to hold a dominant position in both
the wholesale and retail access markets. Once again, the as-efficient competitor standard was applied
to determine the existence of an anticompetitive squeeze.40 This test established that the spread
38 At the time in question, two “baskets” were established, one for services to residential customers and the other for services to business customers. 39 Commission Decision 2003/707/EC of 21 May 2003 relating to a proceeding under Article 82 of the EC Treaty (Case COMP/C-1/37.451, 37.578, 37.579 – Deutsche Telekom AG) (OJ C 263/9, 14.10.2003). 40 See the Deutsche Telekom decision at paragraphs 107-108:
“…there is an abusive margin squeeze if the difference between the retail prices charged by a dominant undertaking and the wholesale prices it charges its competitors for comparable services is negative, or insufficient to cover the product-specific costs to the dominant operator of providing its own retail services on the downstream market…[]...An insufficient spread between a vertically integrated dominant operator’s wholesale and retail charges constitutes anticompetitive conduct especially where
13
between DT’s average retail access price and its wholesale access price was always negative from the
time when local loop unbundling became a legal obligation in Germany in 1998 up to the end of 2001.
Given that DT had sufficient “commercial discretion” to avoid the squeeze by adjusting the retail
prices within its price-capped baskets and in the absence of any objective justification the margin
squeeze amounted to an abuse of dominance by DT. A fine of €12.6 million was imposed for the
violation, which included a lower basic amount to account for the relative novelty of the margin
squeeze calculation and subsequent efforts by DT to reduce the squeeze, plus a 10% reduction to take
account of the regulatory regime as a mitigating circumstance.
DT appealed the 2003 decision to the CFI, where the Commission’s approach was largely affirmed.41
DT further appealed that judgment to the Court of Justice, which in 14 October 2010 dismissed DT’s
action in its entirety.42 Since the Court of Justice judgment essentially confirms the approach of the
CFI, we shall focus on the more recent higher court judgment in determining the principles on margin
squeeze emerging from the case.
Firstly, the Court of Justice accepted that margin squeeze can, in itself, constitute an abuse of
dominance contrary to Article 102 TFEU. The abuse of margin squeeze is concerned with the
unfairness of the spread between two vertically related prices. Therefore, it is not necessary to
establish, in addition, that either the wholesale or retail price is, independently of the claimed squeeze,
excessive (thereby amounting to a constructive refusal to deal) or predatory.43 Moreover, the
unfairness of a margin squeeze stems from its very existence rather than its precise spread. The
absolute values of the wholesale and retail prices are therefore irrelevant: margin squeeze relates
instead to the relationship between these absolute values.44
Secondly, the Court of Justice approved the use of the as-efficient competitor test in order to
determine whether such a pricing spread breaches the competition rules, relying solely upon the
dominant undertaking’s charges and costs, rather than the situation of its actual or potential
competitors. The Court took the view that, not only is such an approach consistent with general
welfare considerations, insofar as it protects competition (in the form of as- or more efficient
competitors) as opposed to less efficient competitors, it is moreover consistent with the general
other providers are excluded from competition on the downstream market even if they are at least as efficient as the established operator.”
41 Case T-271/03 Deutsche Telekom AG v Commission of the European Communities [2008] ECR II-477. 42 Case C-280.08 P Deutsche Telekom Ag v European Commission, Judgment of 14 October 2010 (not yet reported). 43 Deutsche Telekom (CJEU) at paragraph 182. 44 Deutsche Telekom (CJEU) at paragraphs 167-168.
14
principle of legal certainty, insofar as a dominant firm will know its own costs but not those of its
competitors.45
Thirdly, the Court acknowledged that, in certain circumstances including those of the Deutsche
Telekom case itself, the only mechanism available to the dominant firm by which to remedy an
abusive margin squeeze may be to raise retail prices for end-users. Nonetheless, the Court appears to
have taken a “lesser-of-two-evils” approach, whereby increased retail prices are acceptable in the
immediate term in order to protect the existence of competition in the market, on the assumption in
the longer term the presence of competition would place downward pressure on retail prices once
again.46 This position is in complete opposition to the approach of the Supreme Court in LinkLine,
discussed below, where the fear that dominant firms might raise retail prices to avoid a price squeeze
seems to have been a key factor behind the Court’s highly restrictive approach.
Fourthly, the Court followed the approach of the General Court in Deutsche Telekom in holding that,
contrary to the arguments of the Commission, it is not sufficient to merely demonstrate the existence
of a prima facie margin squeeze of as-efficient competitors in order to establish a violation of Article
102 TFEU. Additionally, anticompetitive effect must be established, relating to the possible barriers
which the appellant’s pricing practices could have created for the growth of products on the retail
market, and therefore, on the degree of competition on the downstream market. However, the fact
that a margin squeeze, once implemented, ultimately fails to drive the competitors from the market
does not alter its categorisation as abusive, where the market penetration of those competitors has
been made more difficult.47
Finally, the Court considered the impact that sector-specific regulation has on the application of EU
competition law, and in particular, on the margin squeeze concept. As noted, the case against DT had
proceeded on the assumption that did not have any discretion as to the wholesale access prices to be
charged, which were set by the national regulator, but that it could alter its retail prices provided that
these remained within the total price cap and that the changes were granted regulatory approval. In
assessing the impact of the sectoral regulation, the Court drew a distinction between situations where
the restriction of competition is wholly attributable to the regulatory regime, and situations where it is
45 Deutsche Telekom (CJEU) at paragraphs 196-202. 46 Deutsche Telekom (CJEU) at paragraphs 181-182. The Court did not, however, address the dilemma of the firm that is dominant but subject to price regulation at the wholesale level, and which faces downward competitive prices at the retail level—in particular, the question as to whether such a firm can cut prices to meet competition, even if this leads once again to a margin squeeze. Following the highly restrictive approach adopted in the Telefónica decision, it seems unlikely that a dominant firm can invoke the “meeting competition” defence to excuse a conscious margin squeeze, even though, in a somewhat circular fashion, it is the existence of the prohibition on margin squeeze which, coupled with the wholesale price regulation, has allowed the development of competition which drives the reduction in retail prices (Telefónica at paragraphs 637-640). 47 Deutsche Telekom (CJEU) at paragraphs 250-254.
15
merely encouraged or facilitated by the regulatory regime, also allowing some scope for autonomous
conduct by the firm concerned. Under the former scenario, the competition rules are not applicable,
because they apply directly only to the conduct of undertakings—essentially a State or regulatory
compulsion defence. In the latter case, however, any scope for autonomous behaviour by the relevant
firm can be examined to determine whether it is in conformity with the competition rules.48 A very
high threshold for operation of the State compulsion defence in these circumstances was therefore
established by the Court. The existence of sector-specific regulation—even relatively intrusive
regulation, like the mandatory access requirements and dual-level price regulation in place in the
Deutsche Telekom situation—does not immunise the relevant market from the application of the
competition rules, unless it has “eliminated any possibility of competitive activity” by the dominant
firm.49 As DT had the ability to avoid the margin squeeze by raising its price for retail access, its
failure to do so violated the competition rules. The existence of sectoral regulation is, however,
relevant to the determination of the competitive conditions in the market concerned, a factor that
impacts on the application of the competition rules to the conduct of firms operating in that market.50
It may also lead to a reduction in fine for the firm, where it is found to be a contributing factor to the
breach. The relationship between the margin squeeze concept and sector-specific regulation is
considered further below in Part IV of this paper.
The TeliaSonera Judgment
The principles established in Deutsche Telekom case have been confirmed and somewhat developed
in the recent TeliaSonera judgment of the Court of Justice.51 A reference case from Sweden, it
concerned the application of Article 102 TFEU and the margin squeeze abuse by the Swedish national
competition authority. In the absence of a detailed finding of facts, the referring court asked, in
essence, a series of questions clarifying the parameters of the margin squeeze concept.
The Court of Justice confirmed that margin squeeze constitutes a standalone abuse under Article 102
TFEU, where the spread between wholesale and retail prices is “negative or insufficient…so that that
spread does not allow a competitor which is as efficient as [the dominant] undertaking to compete for
the supply of [retail] services to end users.”52 In such circumstances, an as-efficient competitor of the
dominant firm would be forced to operate at a loss or “artificially reduced levels of profitability.”53
Generally, only the costs and revenues of the dominant firm are relevant to this assessment—an
48 Deutsche Telekom (CJEU) at paragraphs 80-85. 49 Deutsche Telekom (CFI) at paragraph 90. 50 Deutsche Telekom (CJEU) at paragraph 224. 51 Case C-52/09 Konkurrensverket v TeliaSonera Sverige AB, judgment of 17 February 2011, not yet reported. 52 TeliaSonera at paragraphs 31-32. 53 TeliaSonera at paragraph 33.
16
objective standard—although in exceptional circumstances, the costs and prices of its competitors
might be relevant, for example where it is not possible to identify the cost structure of the dominant
firm.54 Proof of recoupment of any losses incurred as a result of the margin squeeze is not required,55
nor is dominance in the downstream market, wholesale dominance being sufficient.56
The Court clarified that the margin squeeze concept can be applied to any pricing practices by
dominant firms, even in the absence of a regulatory or antitrust duty to deal at the wholesale level.
Here, the Court diverged from the advice of Advocate General Mazák, who had argued that unless the
firm has a duty to deal in the first place—either imposed by sectoral regulation or because the
relatively restrictive Oscar Bronner requirements for the imposition of antitrust forced sharing are
satisfied57—it cannot be under a distinct duty to deal on particular terms.58 Instead, the Court was of
the view that where a dominant firm voluntarily chooses to deal with downstream competitors, it
might be found to abuse its dominant position where the terms of dealing are “disadvantageous” to
the latter.59 Indispensability of the wholesale input is not, therefore, required for liability.60
On the other hand, the Court followed the approach in Deutsche Telekom by requiring anticompetitive
effect to be demonstrated in order to establish an abusive margin squeeze. The Court appears to have
established a two-part presumption as regards the effects of a squeeze, based on the level of the
margin between wholesale and resale prices. Where the margin is negative, an effect which is at least
potentially exclusionary is probable. Where the margin remains positive, it has to be demonstrated
that the application of the pricing practice was (for example, by reason of reduced profitability) likely
54 TeliaSonera at paragraphs 41-46. 55 TeliaSonera at paragraphs 96-103. 56 TeliaSonera at paragraphs 84-89. 57 Under the criteria set out by the Court of Justice in Oscar Bronner, three conditions must be satisfied before a refusal to grant access to an input by a dominant firm can be considered abusive: (i) the refusal must be likely to eliminate all competition in the secondary market on the part of the person requesting the access; (ii) there is no objective justification for the refusal; and (iii) the service in itself is indispensible to carrying on that person’s business, inasmuch as there is no actual or potential substitute in existence; see Case C-7/97 Oscar Bronner
GmbH & co.KG v Mediaprint Zeitungs- und Zeitschriftenverlag GmbH & Co.KG [1998] ECR I-7791 at paragraph 41. Note that the Court in TeliaSonera, in a rather curious dicta, seems to suggest that even though the conduct at issue in Oscar Bronner was considered highly unlikely, on the facts, to amount to an anticompetitive refusal to deal, it may have constituted another form of abuse, such as illegal tying (see TeliaSonera at paragraph 57). 58 Opinion of Advocate General Mazák in Case C-52/09 Konkurrensverket v TeliaSonera AB, delivered on 2 September, at paragraphs 11-30. Advocate General Mazák argued, persuasively in my view, that absent a regulatory duty to deal, there is “no independent competitive harm caused by the margin squeeze above and
beyond the harm which would result from a duty-to-deal violation at the wholesale level” (paragraph 16). Moreover, he highlighted the potential dangers of a rule to the contrary, including detrimental impact on a dominant firm’s incentives to invest, as well as the danger that it would most likely remedy a price squeeze simply by raising retail prices (paragraph 21). See also Hovenkamp & Hovenkamp, cited fn. 9 above, at 278. 59 TeliaSonera at paragraph 55. 60 Although raised by the Advocate General in his Opinion, the impact of such a rule on the willingness of firms to deal with their competitors without regulatory compulsion was not considered by the Court.
17
to make market penetration of competitors more difficult.61 Similarly, while indispensability is not
required for liability, it is relevant to the assessment of the likely effects of the squeeze. Thus, where
the wholesale product is indispensable, potential anticompetitive effects are probable.62 Nonetheless,
the Court expressly confirmed that the objective justification defence is available, even where a prima
facie margin squeeze with anticompetitive effects has been established. The dominant firm has the
burden of demonstrating that its pricing practice is economically justified in line with the defence.63
The facts of the TeliaSonera reference did not require the Court to consider the application of the
margin squeeze principle in a regulated sector, and so Deutsche Telekom remains the guiding
authority on this point. Yet, the Court in TeliaSonera construed the “special responsibility” of
dominant firms with respect to margin squeeze so broadly—relating to “the conduct, by commission
or omission, which that undertaking decides on its own initiative to adopt”64—that in regulated
sectors it may develop in practice into a variety of “Bad Samaritan” liability. This point is considered
further in Part IV below.
Other Recent Commission Activity
Deutsche Telekom and TeliaSonera must now be regarding as providing the most authoritative
statement of the status of margin squeeze under EU competition law. However, a number of other
Commission decisions, plus the approach taken in its Communication of 2009 setting out its
enforcement priorities with regard to Article 102 TFEU, are worthy of note.
The Commission’s margin squeeze decision against Telefónica,65 the incumbent telecommunications
operator in Spain, addressed circumstances similar to those of the Deutsche Telekom case. Telefónica
was held to have committed an abusive margin squeeze between its retail and wholesale prices for
high speed internet,66 which were broadly subject to price regulation by a national authority. The
Commission expressly declined to apply the Oscar Bronner test of indispensability67 with respect to
Telefónica’s wholesale products, arguing that this was unnecessary on the facts: there was already in
place a regulatory duty to supply, and Telefónica’s status as a former State monopoly provider meant
61 TeliaSonera at paragraphs 73-74.s 62 TeliaSonera at paragraphs 69-71. 63 TeliaSonera at paragraphs 75-76. 64 TeliaSonera at paragraph 53. 65 Commission Decision of 4 July 2007 relating to a proceeding under Article 83 of the EC Treaty (Case COMP/38.784 – Wanadoo España v Telefónica). 66 Under the market definitions used in the decision, access for high-speed internet services at the wholesale level was divided into regional access, national access and local loop unbundling, which differed as regards the levels of investment required by downstream competitors. The decision itself only concerned regional and national access. Cable internet access, which had coverage in about 40% of Spain at the time covered by the decision, was entirely excluded from the market definition at the wholesale level. 67 See fn. 56 above.
18
that its ex ante incentives to invest in infrastructure were not at stake.68 While this position has been
criticised,69 it is in essence confirmed by TeliaSonera’s view that a regulatory or antitrust duty to
supply is not required for margin squeeze liability. A fine of €151.875 million was imposed on
Telefónica for the breach, which included a reduction of 10% to take account of the impact of the
regulatory regime on its behaviour.70
In the RWE Gas Foreclosure commitment decision,71 the Commission voiced concerns that RWE, the
vertically integrated German gas company, had maintained an anticompetitive margin squeeze
between its gas transmission tariffs and its downstream gas supply tariffs. In addition to the high
level of the network tariffs charged by RWE, important elements of the network tariffs applied only to
third part users, creating an asymmetric effect on network access costs. The case was closed without
a finding of breach, on the basis of a binding commitment from RWE to divest its existing German
gas transmission business. Note that, similarly to Deutsche Telekom, RWE’s obligation to provide
third party access to its gas transmission network had its origins in the EU gas market liberalisation
directives. Given the abridged nature of the commitments procedure, the impact of the regulatory
framework was not considered in the published decision, but it does provide an example of a margin
squeeze occurring in a regulated area other than the telecommunications sector.
Mention should also be made of the guidance issued by the Commission in 2009 on the topic of its
enforcement priorities with respect to the enforcement of Article 102 TFEU.72 In this document, the
Commission itself adopts a noticeably more restrictive approach to margin squeeze that that evinced
in the case law. Margin squeeze is classified as a variety of refusal to deal, and thus is considered
abusive only where: (i) access to the wholesale product is objectively necessary to compete
effectively on a downstream market; (ii) refusal to supply is likely to lead to the elimination of
effective competition on the downstream market; and (iii) the refusal is likely to lead to consumer
harm.73 The extent to which this policy statement will limit the margin squeeze concept in EU
competition law is, however, somewhat doubtful. Firstly, regardless of the Commission’s prior
views, the far more expansive margin squeeze principles outlined by the Court of Justice in its recent
cases must be considered as the authoritative statement of the current EU position. Secondly, the
Commission itself takes the view that, where there is a regulatory duty supply already in place—
68 Telefónica at paragraphs 302-309. 69 See D. Geradin, “Refusal to Supply and Margin Squeeze: A Discussion of Why the “Telefónica exceptions” are Wrong”, available online at ssrn.com/abstract=1750226. 70 Telefónica at paragraphs 765-766. 71 Commission Decision of 18 March 2009 relating to a proceeding under Article 82 of the EC Treaty and Article 54 of the EEA Agreement (Case COMP/39.402 – RWE Gas Foreclosure). 72 Communication from the Commission — Guidance on the Commission's enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings (OJ C 45/7, 24.2.2009). 73 Guidance on enforcement priorities at paragraphs 80-81.
19
meaning, in essence, that any damage to investment incentives has already been done—the higher
refusal to deal standard does not apply. Rather, what must be shown is simply likelihood of
anticompetitive foreclosure.74 Given the frequency with which margin squeeze issues have arisen in
regulated sectors, this may well be the exception that in fact forms the rule.
Margin Squeeze under US Antitrust Law
Although margin squeeze as a standalone antitrust abuse has, arguably, been recognised in US law for
more than half a century since the seminal decision of the Second Circuit in Alcoa,75 the recent
decision of the Supreme Court in LinkLine appears to have eliminated the possibility of maintaining a
independent price squeeze action. In the following section of the paper, we shall consider the Alcoa
and LinkLine jurisprudence, in an effort to determine the scope (if any) for margin squeeze actions
under US antitrust law today.
The Alcoa case
The facts of Alcoa display certain similarities to both the Napier-Brown and Industries des Poudres
Sphériques cases discussed above, and likewise occurred in an unregulated market. Alcoa held a
monopoly in a production of virgin aluminium ingot in the US, which it sold to sheet aluminium
manufacturers, and it also itself produced sheet aluminium. Among a series of monopolisation
offences with which Alcoa was charged was the claim that it “consistently sold ingot at so high a
price that the “sheet rollers”, who were forced to buy from it, could not pay the expenses of “rolling”
the “sheet” and make a living profit out of the price at which Alcoa itself sold “sheet”.”76 Judge
Learned Hand held that, when Alcoa was made aware of the effects of the price squeeze and failed to
remedy it, it became unlawful under §2 of the Sherman Act. Where there was evidence of intent, a
price squeeze in itself could constitute a monopolisation offence:
That it was unlawful to set the price of “sheet” so low and hold the price of ingot so high, seems to us unquestionable, provided…that on this record the price of ingot must be regarded as higher than a “fair price”.77
The Alcoa judgment has attracted significant academic and judicial criticism. This is due, in large
part, to its references to “fair price” and “living profit” for competitors, which are considered to be
inconsistent with the more economic approach to antitrust law adopted by the courts in the decades
74 Guidance on enforcement priorities at paragraph 82. 75 United States v Aluminium Company of America et al. 148 F.2d 416, 65 U.S.P.Q. 6 (1945), commonly known as “Alcoa”. The Court of Appeals for the Second Circuit sat as a court of last instance in this case, as the recusal of a number of Supreme Court justices meant that it was unable to reach the necessary quorum to hear the case, and the principles established in Alcoa have historically be granted the same status as Supreme Court precedent. 76 Alcoa at 437. 77 Alcoa at 438.
20
following the decision. Bork criticised the judgment on the basis that it sought “to trade off consumer
welfare for unarticulated social values.”78 LinkLink relegates Alcoa to a footnote, taking that view
that the more recent (and far more restrictive) precedents in Trinko and Brooke Group were “more
pertinent” in light of “developments in economic theory and antitrust jurisprudence since Alcoa”—
even though neither of those cases, on the facts, dealt with price squeeze issues.79
A key criticism in LinkLine of the Alcoa price squeeze principle is the alleged difficulty of
determining whether a margin is fair or adequate on the facts of a case. However, the test articulated
by Judge Hand in Alcoa in fact amounts to a simplified form of the as-efficient competitor test found
in EU law, also referred to as the “transfer price test” in US case law.80 The administrability of the as-
efficient competitor test in practice may be debatable, particularly in view of the difficulties of
apportioning the costs of a vertically integrated firm;81 nonetheless, there is a clear precedent within
US antitrust for its use. As LinkLine demonstrates, this precedent has not been followed and its status
as good law must be regarded as questionable in the circumstances.
The LinkLine Judgment
The facts of the LinkLine case are remarkably similar to those of Deutsche Telekom and Telefonica,
relating to local loop unbundling under the Telecommunications Act of 1996.82 LinkLine purchased
wholesale DSL transport services from AT&T, the owner of the local fixed telecommunications
network throughout much of California, and both LinkLink and AT&T provided downstream retail
DSL services to end users on the network. AT&T had a regulatory duty to deal with LinkLine,
enforced by the sectoral regulator, the Federal Communications Commission; it does not appear to
have been subject to price regulation at either the wholesale or retail levels. In 2003, LinkLine
brought an antitrust suit against AT&T, alleging inter alia an anticompetitive price squeeze in
violation of §2 of the Sherman Act. AT&T’s motion to dismiss the price squeeze claim was denied
by both the district court and Ninth Circuit of Appeals, and so the case made it to the Supreme Court
in 2009. The principal issue to be decided was whether a price squeeze could constitute a standalone
78 R.H. Bork, The Antitrust Paradox: A Policy at War with Itself (1973) Free Press, New York, USA, p.52. 79 Hovenkamp & Hovenkamp (2009), argue that it is difficult to read LinkLine as anything other than an implicit overruling of Alcoa. 80 See Brief of the American Antitrust Institute as Amicus Curiae in Support of Dismissal of the Writ or
Affirmance in Pacific Bell Telephone Company v LinkLine Communications, at pp.9-10, citing Hovenkamp & Hovenkamp (2009) at 274-275. Note, however, that the Chief Justice Roberts’ majority opinion in LinkLine expressly rejected the “transfer price test” as a means of assessing margin squeeze, on the basis that “it lacks any
grounding in our antitrust jurisprudence”—even though, in substance, it is the same as the test approved by Judge Hand. 81 OECD (2009), cited fn. 4 above. 82 For a highly critical assessment of the performance of the 1996 Act, see Economides (2005).
21
unilateral conduct antitrust offence;83 Chief Justice Roberts, in the majority opinion, held that it could
not.
The judgment is noteworthy in a number of respects. Firstly, the notion of an anticompetitive spread
between wholesale and retail prices, embraced so emphatically by the EU courts, was dismissed
absolutely in LinkLine. The primary reason for rejecting the pricing spread concept appears to be
administrative concerns:
Recognizing price-squeeze claims would require courts simultaneously to police both the wholesale and retail prices to ensure that rival firms are not being squeezed. And courts would be aiming at a moving target, since it is the interaction between these two prices that may result in a squeeze.84
Thus, a price squeeze may violate the antitrust rules only if there is a refusal to deal at the wholesale
level (which could encompass a constructive refusal to deal consisting of excessively high rates or
unduly restrictive or burdensome terms) or predatory pricing at the retail level. An unfair or
inadequate margin in itself is not illegal.
Secondly, the restrictive nature of this holding is reinforced by the highly restrictive precedents relied
upon by the Court to determine the refusal to deal and predatory pricing questions. After Trinko,
refusal to deal by a monopolist will be found to violate the antitrust laws only in the most exceptional
circumstances.85 Similarly, following Brooke Group, a predatory pricing strategy can be impugned
only where it is shown that (i) the prices charged by the monopolist are below an appropriate measure
of its costs, and (ii) there is a dangerous probability that the monopolist will be able to recoup its
“investment” in below-cost prices.86 Moreover, the Supreme Court does not appear to have modified
in any way the test for predation to take account of the cross-subsidisation aspect of margin squeeze,
whereby losses at the retail level can be supported with profits at the wholesale level. As a result, it is
impossible for a margin in itself to breach the antitrust rules, and highly unlikely that either of the
constituent prices will breach these rules either.
83 The rather complicated procedural posture of the case meant that it could in fact have been resolved without reaching any decision as to the viability of price squeeze claims under antitrust law, an approach urged by the American Antitrust Institute in its Amicus Curiae brief in the case on the basis that the absence of a factual record made the case a poor vehicle to establish antitrust policy (at p.5). While Chief Justice Roberts, speaking for the majority of the Court, considered it both possible and necessary to rule on the margin squeeze question, Justice Breyer, delivering a concurring opinion in which Justices Stevens, Souter and Ginsburg joined, limited their holding solely to the regulated firm exception—thus, arguably, suggesting that outside the regulatory context a standalone margin squeeze claim may still be viable; see Hovenkamp & Hovenkamp (2009) at 282. 84 LinkLine, emphasis in original. 85 “[U]ltimately the issue is whether there is a valid business justification for the refusal to deal ot whether the
facts demonstrate a willingness to forgo short-term profits in favour of reducing competition over the long run
by harming a rival”; see Meriwether (2010) at 70. 86 Brooke Group Ltd v Brown & Williamson Tobacco Corp., 509 U.S. 209 (1993) at 222-224.
22
Finally, LinkLine confirmed the approach in Trinko with respect to the effects of sector-specific
regulation on the application of the antitrust rules, and indeed may have strengthened the de facto
antitrust immunity for regulatory activity resulting from that judgment. The Supreme Court in Trinko
took the view that the existence of a regulatory duty to deal removed any scope for imposing an
antitrust duty to deal in the market concerned. Trinko, like LinkLine, concerned the mandatory access
provisions of the Telecommunications Act of 1996, which the Trinko Court described as “a good
candidate for implication of antitrust immunity, to avoid the real possibility of judgments conflicting
with the agency’s regulatory scheme”. Nonetheless, the 1996 Act contains an explicit antitrust saving
clause, which preserves those “claims that satisfy established antitrust standards”. Thus, in Trinko,
the Court had to consider whether an antitrust duty to deal could exist in the circumstances: it took the
view that the extensive provision for access under the 1996 Act made it unnecessary to impose a
judicial doctrine of forced access under the antitrust rules. In LinkLine, by contrast, the Court
interpreted the Trinko precedent as creating a “straightforward” rule, barring any antitrust obligations
arising in the presence of sectoral regulation. Notably, both LinkLine and Trinko relied upon a
judgment of Justice Breyer, given when he was Chief Justice of the First Circuit, in the case of
Concord v Boston Edison.87 That earlier case concerned an alleged price squeeze by a vertically
integrated electricity company, which was subject to price regulation at both the wholesale and retail
levels. In the judgment, Breyer J considered at length the relationship between competition law and
regulation, arguing that “where regulatory and antitrust regimes coexist…antitrust analysis must
sensitively “recognize and reflect the distinctive economic and legal setting” of the regulated industry
to which it applies.”88 On the question of the alleged price squeeze, Breyer J took the view that:
…a price squeeze of the sort at issue here does not ordinarily violate Sherman Act, §2, where the defendant’s prices are regulated at both the primary and secondary levels. In so holding, we are not saying either that the antitrust laws do not apply in this regulatory context, or that they somehow apply less stringently here than elsewhere. Rather, we are saying that, in light of the regulatory rules, constraints, and practices, the price squeeze at issue here is not ordinarily exclusionary, and, for that reason, it does not violate the Sherman Act.89
Note how the approach of the court in Town of Concord to the question of the antitrust/regulation
interface differs markedly from the approach in Linkline, even though the Supreme Court in the latter
decision quoted at length from the earlier decision. In Town of Concord, antitrust is applied within
the regulated sector, but the existence of the regulation so impacts upon the relevant market that the
conduct at issue does not breach the antitrust rules. By contrast, in LinkLine this ultimate result is
presumed, so that a per se rule of legality under the antitrust rules is imposed, albeit this is conceived
87 Town of Concord, Massachusetts, et al. v Boston Edison Company 915 F.2d 17 (1990). 88 Town of Concord at paragraph 20, quoting Watson & Brunner, “Monopolization by Regulated “Monopolies”: The Search for Substantive Standards” (1977) 22 Antitrust Bulletin 559, at 565. 89 Town of Concord at paragraph 21. See also paragraph 35: “Full price regulation dramatically alters the
calculus of antitrust harms and benefits.”
23
both in Trinko and LinkLine as more of an immunity than a safe harbour. Nonetheless, the ultimate
result in Town of Concord, Trinko and LinkLine is the same: the presence of economic regulation
precludes any finding of breach of the antitrust laws. This result stands in clear contrast to the
position in EU law, as confirmed in Deutsche Telekom and TeliaSonera, and it is this question—the
application of the margin squeeze concept in regulated sectors—that comprises Part IV of this paper.
Before addressing that issue, however, it is worth summarising the differing approaches to margin
squeeze to be found under current EU and US jurisprudence.
- In the EU, a margin squeeze can constitute a standalone violation of Article 102 TFEU where the
spread between wholesale and retail prices charged by a vertically integrated firm is negative or
insufficient to cover the costs incurred by that firm (or its as-efficient competitor) in producing
the retail product. There must be dominance at the wholesale but not the retail level and, barring
exceptional circumstances, the only costs and revenues to focus on are those of the dominant firm
itself. It is necessary to demonstrate that the existence of the squeeze makes market penetration
more difficult for competitors, although where the wholesale product is indispensable and/or the
spread is negative, anticompetitive effects are considered to be probable.
- In the US, margin squeeze does not constitute a standalone violation of §2 of the Sherman Act.
Price squeezing behaviour must instead be assessed as a constructive refusal to deal or an instance
of predatory pricing—with no modification of the existing (and highly restrictive) legal tests for
these offences to take account of the vertical relationship and possibility of cross-subsidies.
- In the EU, the presence of sector-specific regulation—even intrusive regulation that mandates
entry and sets prices—does not prevent the application of competition law and the margin squeeze
concept, provided the vertically integrated firm retained some scope to avoid the squeeze, even if
it can only do so by raising retail prices. By contrast, in the US, the ex ante economic regulation
of a sector appears to remove it from the purview of antitrust, so that only regulatory duties can
arise and regulatory remedies be imposed.
IV. Margin Squeeze and Economic Regulation
We now consider in greater detail the relationship between the margin squeeze concept and sector-
specific economic regulation. Both competition law and economic regulation are concerned, broadly,
with the effective functioning of markets. However, important differences remain—for example, the
24
mode and temporal aspect of enforcement, as well as the objectives that may be pursued90—so that
one regime cannot be subsumed within the other without reaching a very high level of generalisation.
Economic regulation can take a multitude of forms,91 but general, the three key decision variables
controlled are price, quantity and industry entry or exit.92 Moreover, regulation of one aspect of a
firm’s operations frequently impacts on the other variables.93 For example, mandatory access
requirements almost inevitably lead to some form of price regulation for the industry concerned,
because a right of access or supply will be worthless in practice if the infrastructure owner or supplier
is permitted to charge an unreasonably high rate that amounts to a constructive refusal to deal with
competitors. Similarly, production quotas typically act as a barrier to entry, since potential entrants
may be prevented from bringing their product onto the market. In the margin squeeze context, the
problem is frequently created by a mandatory access requirement at the wholesale level—and
complicated by price regulation at one or both levels of the price squeeze.
Three preliminary observations are appropriate. There is, firstly, an asymmetry of liability regarding
margin squeezes that occur in a regulated sector. In such circumstances, the margin squeeze cannot
be caused entirely by the conduct of the vertically regulated firm. By contrast, the squeeze can be
caused entirely by the regulator.94 Secondly, regulated markets are typically very far from the
archetype of atomistic perfect competition that is the model informing the competition rules. Indeed,
a margin squeeze in a regulated sector may in fact be evidence of increased competition in the market
(due to intermodal competition), which in fact may imply that deregulation is appropriate.95 Thirdly,
and in view of the rather critical approach that this paper takes with respect to the regulated margin
squeeze as a competition law offence, it should be noted that the concurrent application of
competition law and regulation is not without its strong defenders.96 Indeed, it has been argued that
the presence of regulation increases rather than reduces the need for competition law enforcement,
90 See Geradin & O’Donoghue (2005) at 360-364, for an outline of the basic differences between regulatory and competition law powers in relation to margin squeeze. 91 See Viscusi et al. (2005), Part II, for a comprehensive analysis of the wide variety of forms of economic regulation that may be implementation and the particular features of each. 92 Viscusi et al., (2005), at p.358. 93 For example, if the regulator imposes a quota restricting the permitted output of a particular product, in the absence of price regulation the artificial scarcity is likely to result in a rise in prices for that product. Moreover, if the quota restricts output per firm but not total industry output, the rise in prices is likely to encourage new entry. Conversely, if the quota caps total industry output, it will act as a barrier to entry as potential new entrants have to The EU’s Common Agricultural Policy provides an (in)famous example of how regulation can impact on every aspect of a sector. 94 Lopatka (1984) at 597. 95 W. Briglauer, G. Götz & A. Schwarz, “Can a Margin Squeeze Indicate the Need for Deregulation? The Case of Fixed Network Voice Telephony Markets” (2010) 34 Telecommunications Policy 551. See also Sidak (2008) at 300, who argues that a prima facie margin squeeze may reflect merely the efficient workings of the market and the effective response to consumer demand. 96 See, for example, Dogan & Lemley (2009); also A. Heimler, “Is Margin Squeeze an Antitrust or a Regulatory Violation” (2010) 6(4) Journal of Competition Law & Economics 879.
25
insofar as it creates incentives for the regulated (and usually, former monopoly) firm to avoid the
regulation and seek monopoly rents in new markets.97
The US Approach to the Regulated Margin Squeeze
Under US jurisprudence, the existence of regulation at one or both levels appears to remove that
element of the squeeze from the purview of competition law. While the extent to which Trinko in fact
removes regulated conduct from the purview of competition law is disputed,98 LinkLine articulates a
clear principle that regulation prevents the relevant prices from being examined even individually
under the discrete tests for refusal to deal or predation.
The question of whether margin squeeze behaviour should be exempt from the application of the
competition rules where it is also subject to sector-specific regulation is linked to the broader issue of
the intersection between competition law and regulation. The regulatory immunity approach adopted
in LinkLine has the benefit of clear, ex ante certainty of application for firms and for national
authorities (regulators and competition agencies). Moreover, a key assumption of this approach is
that the competition problem is not irremediable absolutely, insofar as a regulatory remedy generally
exists—it is simply irremediable via antitrust. In both Trinko and LinkLine, a regulatory remedy was
available to address the competition problem, and indeed in Trinko, the incumbent’s behaviour had
already been investigated and corrected by the sectoral regulator, the FCC. The availability of
regulatory remedies appears to have been a significant factor confirming the Supreme Court decision
in both cases.
Of course, where the regulatory regime does not have the means by which to remedy the margin
squeeze, excluding the application of competition law in order to remedy the problem creates a lacuna
in the legal framework. The question then becomes whether it is appropriate to address regulatory
problems via competition law mechanisms. Is the application of the competition rules really a
suitable vehicle by which to tackle the problem of regulatory capture, for example, to say nothing of
adverse effects on legal certainty that may result? A key policy question is whether we accept that
some competition problems are beyond the reach of competition law—and much comes down to the
administrability of competition law in the particular regulatory context.99
97 T.J. Brennan, “Essential Facilities and Trinko: Should Antitrust and Regulation be Combined?” 61 Federal
Communications Law Journal 133, at 141. 98 It has been argued, for example, that while Trinko does not, a priori, remove a regulated sector from the purview of antitrust, the pre-existence of the regulatory regime means that the conduct at issue is unlikely to be considered harmful on the facts; see J.L Rubin, “The Truth About Trinko” (2005) 50 Antitrust Bulletin 725. By contrast, Meriwether (2010), at 70, argues that it leaves “little room” for competition enforcement unless the regulatory regime itself is inadequate to address the harm. 99 Hovenkamp & Hovenkamp (2009) at 288.
26
A further question with regard to the regulatory immunity approach is whether immunity should be
automatic—as was apparently the case in LinkLine—or reached via a deductive reasoning process that
considers the impact of the regulatory regime on the firm’s behaviour, and in particular, its potential
to cause anticompetitive harm—as was laid out in Town of Concord. The former approach has the
benefit of administrative simplicity, while the latter is arguably the more principled and respectful of
the jurisdiction of the antitrust rules. To an extent, however, given that both approaches will arrive at
the same ultimate conclusion, this is perhaps more of a theoretical question than a practical concern.
The EU Approach to the Regulated Margin Squeeze
In the EU, by contrast, the concurrent application of competition and regulation has been expressly
approved by both the Commission and the EU courts, and this principle extends to the regulated
margin squeeze context. The Commission is strongly of the view that competition rules and sector-
specific regulation form “a coherent regulatory framework” of EU law,100 capable of concurrent
application and with “mutually reinforcing” effect.101 In Deutsche Telekom, the Court of Justice
agreed, describing the relationship between the EU competition rules and sector-specific regulation in
the following terms:
…the competition rules laid down by the EC Treaty [now Treaty on the Functioning of the European Union] supplement…by an ex post review, the legislative framework adopted by the Union legislature for ex ante regulation of the telecommunications markets.102
Advocate General Mazák, in his Opinion, in Deutsche Telekom, saw the sector-specific regulation at
issue in that case and the EU competition law provisions as “complementary”,103 arguing that the
former “completes the Treaty provisions on competition and should guarantee a competitive context
to an extent which [the competition rules] alone could not with the same certainty…[the competition
rules] should be considered to constitute a set of minimum criteria.”104 Competition law and
regulation remain separate instruments, “even if ultimately they both seek to promote competition”:
100 Commission Notice on the application of the competition rules to access agreements in the
telecommunications sector (C 265/2, 22.8.98) (“Access Notice”), at paragraph 57. While framed in terms of telecommunications markets, the principles contained in the Access Notice is considered to be of more general relevance (see paragraph 6). 101 Access Notice at paragraph 58. 102 Deutsche Telekom (CJEU) at paragraph 92. Note that the Commission in the Telefónica decision, discussed below, took the ex ante/ex post distinction further: it declined to use to the costs relied upon by the national regulator to set access prices ex ante, on the basis that costs assessed ex post are more accurate (but, self-evidently, unavailable to the regulator at the price-setting stage); see Telefónica at paragraphs 492-511. 103 Opinion of Advocate General Mazák in Case C-280/08 P Deutsche Telekom AG v European Commission, delivered on 22 April 2010, at paragraph 19. 104 Opinion in Deutsche Telekom at paragraph 15.
27
here, Advocate General Mazák invoked the analogy of two barriers, the distinct requirements of each
having to be respected on the facts.105
In general terms, there is, first and foremost, a basic legal certainty and fairness problem stemming
from concurrent application of regulation and competition law. O’Donoghue argues that the approach
in Deutsche Telekom shows “a lack of coordinated thinking with respect to the interaction between
regulation and competition law and, as a result, [has] placed regulated firms in an invidious
position.”106 Thus, the regulated firm is subject to two sets of legal duties, both imposed by the State,
with respect to the same market conduct, and compliance with one set of obligations does not
guarantee that its other obligations have also been satisfied, as the facts of Deutsche Telekom itself
illustrate. At a more pragmatic level, there is a risk of duplication of work by enforcers107—which,
incidentally, appears to be the Commission’s primary concern regarding concurrent
regulation/competition law enforcement.108
More specifically, in the margin squeeze context, the fact of regulatory price setting means that the
prices under assessment have not been freely decided upon and implemented by the dominant firm.
Thus, liability is imposed on the firm, not for its anticompetitive actions, but rather for the
anticompetitive consequences of its inaction. It is established in EU law that a dominant firm has a
“special responsibility” not to allow its conduct to impair genuine undistorted competition on the
Common Market.109 In Deutsche Telekom, the Court of Justice distinguished between “scope” to
remedy a margin squeeze (where a dominant firm can, directly or indirectly via representations to the
regulatory price-setter, alter at least one of the parameter of the vertical margin to alleviate the
squeeze) and “fault” in failing to exercise such scope to remedy, for example intention to foreclose as
a result. Once scope has been established, under current EU law there is no additional requirement to
demonstrate that the firm was at fault in failing to exercise its ability to remedy—the mere omission to
do so is sufficient to establish margin squeeze liability.110 TeliaSonera now confirms that a margin
squeeze can be carried out by “commission or omission”.111
A basic issue here is the degree to which a firm subject to detailed regulatory requirements in reality
has scope to proactively avoid the margin squeeze created by a regulated pricing structure. In
Deutsche Telekom, for example, the Court of Justice took the view that the regulation merely
105 Opinion in Deutsche Telekom at paragraph 21. 106 R. O’Donoghue, “Regulated the Regulated: Deutsche Telekom v European Commission” Global Competition
Policy, May-08 (1), at 14. 107 Geradin & O’Donoghue (2005) at 409. 108 Access Notice at paragraphs 28, 150. 109 Deutsche Telekom (CJEU) at paragraphs 83 & 176. 110 Applying Deutsche Telekom (CJEU) at paragraphs 89, 113 & 124. 111 TeliaSonera at paragraph 53.
28
encouraged the squeeze identified. In such circumstances, the dominant firm is expected “to take the
initiative” and to apply for variation of its regulated prices.112 Consider, however, that the regulation
at issue involved (i) a mandatory duty to deal; (ii) wholesale prices that were (according to the
Commission’s case theory) wholly set by the regulator; and (iii) retail prices that were subject to a
price cap that was steadily decreased over the time period concerned, and in any event retail price
changes required the permission of the regulator. To say that DT had genuine scope to alter its prices
to avoid the scope in these circumstances sets the threshold for a finding of autonomous conduct at a
nonsensically low level. At most, DT had a highly circumscribed ability to alter its retail prices, and
therefore it could touch only one parameter of the anticompetitive spread, a term which on its
common sense means implies the need for at least two references points. Yet, the Court refused to
consider the legality of DT’s retail prices in isolation.113
A deeper problem is that, to hold a dominant firm liable for its omission or failure to lobby for
changes to the regulatory regime that will benefit solely the balance sheet of its competitors, is in
effect to establish a “Bad Samaritan” rule, or duty to rescue, in EU competition law. That is, a firm
may have breached the competition rules in circumstances where the anticompetitive effects felt on
the market result from the actions of a third party, namely the regulatory authority that has either set
or approved the prices for the products concerned, thereby dictating in law the prices that the
dominant firm must charge. Thus, the dominant firm will be liable, not for its own conduct—that is,
charging prices required by law—but rather, for its failure to intervene and prevent the regulator from
adopting policies that harm the dominant firm’s competitor.114
112 Telefónica at paragraph 672. 113 While on the facts, DT access retail access prices may have been priced below cost (but not its prices for retail operations viewed as a whole), there would be a strong argument for a State compulsion defence here on the facts. 114 Imposition of liability for Bad Samaritans, who fail to rescue third parties, adheres in large part to the civil/common law divide. Typically, there is a duty to rescue in civil law system, whereas there is no such duty under the common law. For discussion on the role played by the Bad Samaritan doctrine in civil law jurisdictions, see, inter alia, F.J.M. Feldbrugge, “Good and Bad Samaritans: A Comparative Survey of Criminal Law Provisions Concerning Failure to Rescue” (1966) 14 American Journal of Comparative Law 630-657; M. Vranken, “Duty to Rescue in Civil Law and Common Law: Les Extrêmes se Touchent?” (1998) 47 International and Comparative Law Quarterly 934-942; E.A. Tomlinson, “The French Experience with Duty to Rescue: A Dubious Case for Criminal Enforcement” (2000) 20 New York Law School Journal of International & Comparative Law 451-499; D. Schiff, “Samaritans: Good, Bad and Ugly: A Comparative Law Analysis” (2005) 11 Roger Williams University Law Review 77-141; J. Dopico Gómez-Aller “Criminal Omissions: A European Perspective” (2008) 11 New Criminal Law Review 419-451; and J-M. Sílva Sánchez, “Criminal Omissions: Some Relevant Distinctions” (2008) 11 New Criminal Law Review 452-469. For a discussion of the place of the Bad Samaritan doctrine in English law, see T. Elliott & D. Ormerod, “Acts and Omissions – A Distinction Without a Defence?” (2008) 39 Cambrian Law Review 40-59. Arguments that the common law should incorporate some form of duty to rescue can be found at E.J. Weinrib, “The Case for a Duty to Rescue” (1980) 90(2) Yale Law Journal 247-293, and K. Levy, “Killing, Letting Die, and the Case for Mildly Punishing Bad Samaritanism” (2010) 44(3) Georgia Law Review 607-695. See also J. Feinberg, “The Moral and Legal Responsibility of the Bad Samaritan” (1984) 3 Criminal Justice Ethics 56-69 and K. Huigens, “Virtue and Inculpation” (1995) 108 Harvard Law Review 1423-1480 for assessment of the Bad Samaritan doctrine from a moral/ethical perspective.
29
At least two arguments can be advanced against imposing Bad Samaritan liability in the context of a
regulated margin squeeze. The first is that, even in civil law jurisdictions, liability of this nature is
typically imposed only in circumstances where the victim is in imminent danger of serious physical
harm. In the regulated margin squeeze context, the harm is diminished profitability or at worst
bankruptcy for the competing firm; the harm occurs more gradually, usually over numerous or even
years; and during this period, the competing firm has the opportunity to avoid or mitigate the harm by
altering its business strategy or itself petitioning the regulator to impose more appropriate wholesale
and/or retail prices. The duty to rescue presupposes that, without the rescuer’s intervention, the harm
will occur, whereas this crucial causal link is missing in the regulated margin squeeze context. Thus,
the key normative considerations justifying imposition of liability for omission—serious physical
harm, lack of alternative options for avoidance, the constrained time frame—are not satisfied with
respect to a regulated margin squeeze.115
Secondly, the foundational premise of the market system—and thus of competition law, which
protects the market system—is that individual market operators are rational self-maximisers.116 Given
that the welfare of society is merely the sum of the welfare of each individual (or firm), economics
tells us that, somewhat counter-intuitively, when we act in our own best interests we actually act in
the best interests of society. While it would go too far to suggest that competition law advocates
selfishness, it does work from the premise that individual firms should be concerned solely with
securing their own economic efficiency and success in the market. For this reason, dominant firms
are prohibited from using their market power to exclude competitors by recourse to non-normal
means of competition. Rather, they can ensure their continued economic success only by competing
effectively in the marketplace. A rule that requires a dominant firm to consider, and pro-actively
protect, the market position of its competitors goes against this fundamental distinction between
competition on the merits and artificial manipulation of the market that lies at the core of competition
law theory. If dominant firms are prohibited from interfering in the market to the detriment of
competitors, why are they obliged to do so in order to protect their competitors? Surely, the success
or otherwise of competing firms should concern the dominant firm only indirectly if at all, insofar as
115 To give a rather lurid example of this distinction, under a Bad Samaritan law, a passer-by may be liable if she fails to render first aid or call the emergency services upon encountering the victim of a heart-attack, even though the passer-by had nothing to do with the initial occurrence of the heart-attack. Conversely, even where the heart-attack is a direct result of years of eating unhealthy food prepared by a loved one, the cook should not be held responsible for the resulting harm, because the victim’s own actions and inactions were the key causal element. 116 This assumption has been the basis of neo-classical economic theory since the time of Adam Smith, who used the oft-quoted metaphor of the “invisible hand” of the market process, which puts the resources of society to the best (most efficient) uses; see A. Smith, The Wealth Of Nations (1776) Harriman House Edition, republished 2007, Harriman House, Hampshire. For a more recent re-statement of this approach, see G.S. Becker, The Economic Approach to Human Behaviour (1976) University of Chicago Press, Chicago, US.
30
it provides a measure by which to assess the effectiveness of the dominant firm’s own competition
efforts. Despite suggestions to the contrary,117 the Commission itself takes the view that in applying
EU competition law, “what really matters is protecting an effective competitive process and not
simply protecting competitors.”118 Yet, a principle to the effect that a dominant firm must ensure that
third parties act in the best interest of its competitors (with the corollary that its own interests are
likely to be compromised) makes the dominant firm, in effect, its competitors’ keeper. Moreover, it
conflicts with the principle that competition—unlike sector-specific regulation—does not impose
affirmative duties even on dominant firms.119 As for the argument that protecting competitors is one
means of protecting the competitive process, Geradin & O’Donoghue argue that, while subsidising
inefficient entry in the short term on the basis that the entrant will become more efficient over time
may be a legitimate objective under regulatory policy, no such mandate exists under competition
law.120
The Deutsche Telekom case itself highlights a further problem that can result from a conglomerate
regulation/antitrust approach in the EU context. On the facts of the case, DT’s own provision of retail
access services was in fact loss-making; however, losses due to the below-cost access charges were
subsidised by higher call charges. This pricing structure was a result of the historical pricing policy
pursued by DT in its guise as the State monopoly telecommunications provider, on social policy
grounds (presumably, to make access affordable, even if service usage costs were high), which DT
inherited upon privatisation. That telecommunications tariffs had not yet been rebalanced to a more
cost-reflective structure stemmed, in reality, from the failure of Germany to implement correctly the
requirements of the EU telecommunications directives.121 In applying the margin squeeze test,
however, the Commission proceeded as if tariffs had already been rebalanced by refusing to take
account of the cross-subsidisation effects from profitable service charges (which both DT and its
competitors used to make up the access shortfall). Moreover, DT then duly failed the test, precisely
117 See Brief of Amici Curiae Professors and Scholars in Law and Economics in Support of the Petitioners in Pacific Bell Telephone Company v LinkLine Communications at p.5:
The alternative to consumer-welfare maximisation is the view that antitrust law is simply one more tool
of industrial policy, and thus its application may permissibly compromise consumer welfare to advance
the welfare of competitors. Other nations evidently consider this normative proposition to be
appropriate, if recent developments in the European Union are a valid indication. More than ever
before, the United States and Europe appear to be at a fork in the road over whether the law of
monopolization exists to protect consumers of to ensure that a specified number of firms will profitably
populate the market. 118 Communication from the Commission, Guidance on the Commission’s enforcement priorities in applying
Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings (OJ C 45/7, 24.2.2009), at paragraph 6. 119 Geradin & O’Donoghue (2005) at 364. 120 Geradin & O’Donoghue (2005) at 395. 121 Deutsche Telekom (CFI) at paragraph 260.
31
because tariff rebalancing was not yet in place.122 Therefore, by the Commission’s choice to pursue
DT under the competition rules for the margin squeeze abuse rather than Germany for failure to fulfil
its EU law obligations under Article 258 TFEU (ex Article 226 EC), arguably the end result is a form
of horizontal direct effect of the relevant directives, contrary to the established principle of EU law
prohibiting such a result.123 Note that while Geradin & O’Donoghue have highlighted the “arguably
greater scope (and need) for intervention on the basis of competition rules in the [EU] than in the
US” in the context of telecommunications law, on the basis that the latter legislative regime is far
more detailed and prescriptive than the EU regulatory framework,124 this distinction does not address
the legal issue of horizontal direct effect, even though it may provide a policy reason supporting
concurrent application of competition law with the regulatory provisions.
V. Conclusion
Margin squeeze as a competition law abuse addresses the anticompetitive margin between wholesale
access prices and end user retail prices charged by a dominant vertically integrated firm. It is
recognised as a discrete offence under EU competition law, but can be attacked under US antitrust law
only by demonstrating that one or both of the price levels, individually, is anticompetitive. Margin
squeezes arising in regulated sectors pose a particular problem: to what extent is the pricing behaviour
attributable to the regulatory regime rather than the dominant firm’s conduct, and does this disconnect
shield the dominant firm from liability under the competition rules?
Formally, at least, neither EU nor US law contains a bright line rule regarding the application of the
margin squeeze concept within regulated sectors. Nonetheless, after Deutsche Telekom, the scope for
reliance upon the State compulsion defence seems minimal, while in LinkLine antitrust immunity was
treated as de facto automatic in the presence of sectoral regulation.
In a sense, however, the distinction between the EU and US approaches can be overstated. Both are
concerned primarily with preserving competition on the merits in a market, and conversely, avoiding
conduct that inhibits such beneficial competition. Where these positions differ is the focus of their
attentions. From the EU perspective, protecting competition means attracting and keeping rivals (that
is, firms other than the vertically integrated firm) in the market.125 From the US perspective, this
122 The tariff rebalancing requirements under EU telecommunications law, and the application of these principles to the facts of Deutsche Telekom, are discussed in particular detail in the General Court’s judgment in that case. See also Advocate General Mazák’s Opinion in the case at paragraphs 54-59. 123 Case 152/84 Marshall v Southampton and South-West Hampshire Area Health Authority (Teaching) [1986] ECR 723 is the classic authority for the vertical/horizontal direct effect distinction with respect to directives. 124 Geradin & O’Donoghue (2005) at 417. 125 The Commission takes the view that a diminution in competition—either because competitors have been excluded or they compete less vigorously as a result of a margin squeeze—“inevitably leads to likely harm to
32
means protecting the incentives that the vertically integrated firm itself has to compete, for example
by investing in more efficient technology and cutting its retail prices. As to whether, from an
economic standpoint, one approach is superior to the other, it is difficult to say with much certainty.
Given the strong industrial policy flavour that permeates the Deutsche Telekom and Telefónica
decisions—the key objective of development of the information society in the EU—there is perhaps
an argument that the margin squeeze concept developed in that line of case law should be confined to
its facts.126 However, the TeliaSonera case, even though it concerns an alleged margin squeeze in the
same (telecommunications) sector, draws the parameters of the margin squeeze concept in such a
broad manner, and of such general application, that it appears to limit the scope for any such
restrictive interpret. For the moment, therefore, the margin squeeze concept under EU competition is
far from a principle in danger of being narrowed out of existence.
As for the US approach, given contemporary policy arguments in favour of reregulation (or at least
better regulation) in many sectors, the question is whether the automatic regulatory immunity afforded
in LinkLine will be challenged going forward, given that the expansion of economic regulation will
result in a correlating elimination of antitrust. In both jurisdictions, therefore, we must await further
developments before the scope of the margin squeeze concept can be determined definitely.
consumers” (Telefónica at paragraph 557). See also Telefónica at paragraph 586: “The establishment of
foreclosure effects does not mean that rivals are forced to exit the market: it is sufficient that rivals are
disadvantaged and consequently led to compete less aggressively.” 126 See also C. Cavaleri Rudaz, “Did Trinko Really Kill Antitrust Price Squeeze Claims? A Critical Approach to the LinkLine Decision through a Comparison of E.U. and U.S. Case Law” (2010) 43 Vanderbilt Journal of
Transnational Law 1077, at 1119.